Testing the effectiveness of a fabric boom in inproving water quality at a bathing beach.
Most of the beach closings and advisories were related to high levels of bacteria in the water. In some cases, departments of health close beaches pre-emptively based on an evaluation of empirical data both current and historical. For example, the Westchester County Department of Health has a policy of closing all beaches in Mamaroneck Harbor when there is more than one-half inch of rainfall in a 24-hour period. This policy is based on 15 years of sampling data that show a correlation between high levels of coliform and fecal coliform in runoff water entering the harbor and in harbor beach water and one-half inch or more of rain.
The Village of Mamaroneck, with a population of more than 17,000, located in southern Westchester County, New York, operates one of the beaches in Mamaroneck Harbor. In the spring and summer of 1992, the Village sought to determine if it could improve the bacteriologic quality of the water in its bathing beach in Mamaroneck Harbor. Westchester County, with a population of just under 875,000, is bound entirely on the west by the Hudson River and on the south by Long Island Sound. It has over 10 miles of beach on Long Island Sound. Mamaroneck Harbor empties into and receives water from Long Island Sound. The harbor also receives fresh water from two major watersheds: Beaver Swamp Brook and Mamaroneck River. These two river basins cover more than 18,000 acres of Westchester County. Water samples collected from these water courses have frequently revealed high levels of coliform bacteria, and these water courses are considered a significant source of bacteriological contamination within the harbor.
The quality of water at the Mamaroneck Village Beach has historically been marginal, at best. For example, according to a March 1991 U.S. Environmental Protection Agency (EPA) report on Mamaroneck Harbor, between 1983 and 1989, the geometric mean for total coliform at this site was 2,700 Most Probable Number per 100 milliliters (MPN/100 mL). The geometric mean for fecal coliform during this period was calculated at 850 MPN/100 mL (4). Each of these averages exceeded the New York State Department of Health bathing standards (2,400 MPN/100 mL for total coliform and 200 MPN/100 mL for fecal coliform) (5). In 1992, the WCDH closed Mamaroneck Village beach for nine time periods, each for one or more days, due to excessive rainfall.
In an attempt to determine if it was possible to protect the enclosed beach waters from contaminated fresh water running through Mamaroneck Harbor and into the sound, a "fabric boom" was installed at the Village's beach on June 11, 1992. This boom consisted of a large (16-inch diameter) styrofoam ring, over which was draped a 15-foot curtain of porous non-woven propylene fabric. The average pore size is 20 microns.
The boom extended down and was secured to the sea floor. In Mamaroneck this was approximately 15 feet at the deepest point. The boom was secured at the sea floor every 25 feet by a pick type anchor that digs into the sand. It was weighted with a chain anchored at several sites from one end of the beach above the high tide mark into the water in a horseshoe shape, and back to the beach ending at another point above the high tide mark. The installed fabric boom defined the perimeter of the bathing beach waters.
Although the boom is designed to filter out a variety of pollutants and debris, the primary interest was its ability to reduce the number of enteric microorganisms entering the enclosed beach waters. Analysis sought to examine the effectiveness, as measured by total and fecal coliform bacteria, of the installation of the fabric boom in improving the bathing beach water quality during the 1992 summer swimming season at Mamaroneck Village Beach. Chlorides were also measured to determine the level of salinity within and without the boom, and, thereby, to evaluate the flow of water through the boom.
A similar fabric boom was used in the summer of 1990 at Sea Cliff Village Beach in Nassau County, New York, in an effort to reduce coliform bacteria. The Nassau County Department of Health concluded that "the curtain boom appears to have contributed to reducing coliform bacteria levels within the perimeter" (6). Water sample collection for bacteriological and chloride analysis
Prior to the installation of the boom, Westchester County Department of Health (WCDH)/Bureau of Public Health Protection (BPHP) staff collected four water samples on each of eight days from May 25 to June 10, 1992. The samples were collected from water areas that approximately corresponded to Sample Sites 2 and 3, which would subsequently be inside the boom, and Sample Sites 1 and 4, which would be outside of the boom. No samples were collected from Sites 5 and 6 prior to boom installation.
After installation of the boom, WCDH/ BPHP staff collected, or supervised the collection of, four to six site-specific water samples at the Mamaroneck Village Beach. Samples collected at the sites closest to the shore were from water which was approximately 2 feet deep.
The deep water samples (Sites 5 and 6) were collected by the beach lifeguards on an intermittent basis. WCDH/BPHP staff gave the sampling container to the life-guards, instructed them on how and where to collect the sample, and then observed individual guards as he or she collected the sample and returned it to the staff person.
WCDH/BPHP staff collected samples at the shallow sites by inserting a closed sampling bottle under approximately 6 inches of water and then opening it so as to collect a representative sample of water; that is, not just surface water. All samples were collected within a minimum of 2 feet in either direction from the boom. Following boom installation, Sample Sites 2, 3 and 5 were located inside the boom and Sample Sites 1, 4 and 6 were located outside the boom. A total of 178 samples were collected and analyzed for bacteriological content. For total chlorides, water samples were collected from Sites 1, 2, 3 and 4 from June 22 through August 19, 1992; and intermittent sampling occurred at Sites 5 and 6 during the same period. The chloride samples were collected using the same collection procedure as the bacteriological samples. A total of 155 samples were collected and analyzed for chlorides.
Water sample laboratory analysis
The Westchester County Public Health Laboratory (WCPHL), which is certified by the New York State Environmental Laboratory Approval Program (ELAP) and the FDA for examination of coliform bacteria in estuarine waters, performed all sample analysis. The samples were collected in approved containers provided by WCPHL and transported to the laboratory in thermal carrying cases. The cases maintained the water samples at a temperature below 4 |degrees~ C. The WCPHL performed total coliform, fecal coliform, and total chlorides tests. For total and fecal coliform levels the laboratory used the Most Probable Number (MPN) method. For chlorides, EPA Method 325.3 -- Titrimetric Method, which utilizes mercuric nitrate, was used.
To compare water sample results, total coliform and fecal coliform bacteria counts of all samples were converted to logarithms. The mean of the logarithms was determined, and the mean logarithm was converted back to a whole number. These whole numbers -- the means of sample logarithms -- were compared for each of the sampling sites. This arithmetic process is a standard method TABULAR DATA OMITTED used to give less weight to individual samples which are extremely high or low, and yield a true value that is more representative of the mean. This process was also followed for all chloride samples.
To determine if differences in sample results were due to chance, matched pair statistics were employed. T-tests were performed to determine the differences between the values that were observed inside the boom and those observed outside the boom. To determine statistical significance a p value of |is less than~ .05 was chosen.
Coliform samples prior to installation of the boom
Table 1 presents comparisons of the sample results from Sample Sites 1 through 4 before the boom was installed. Samples were collected at sites "inside" and "outside" where the boom was to be installed later in the season. (Table 1a provides measurements for each date and site.) No data was collected or reported for Sites 5 and 6, which are the sites furthest from the beach.
A comparison of the means of the total coliform sample logs for Site 2 vs. Site 1, and for Site 3 vs. Site 4 revealed no statistically significant differences (t=1.50 and t=.35, respectively) in the quality of the water as measured by total coliform, at these sites prior to the boom's installation.
Similarly, no statistically significant differences were found between fecal coliform levels for Site 2 vs. Site 1 and Site 3 vs. Site 4 (t=2.03 and t=.35, respectively). When sample results from "inside" sites were combined and compared with those of combined "outside" sites, again there was no statistically significant difference between them (t=.89 for total coliform and t=.83 for fecal coliform).
Coliform samples following installation of the boom
Table 2 presents the mean of sample logs for each site and for the combination of sites inside and outside the boom, following installation of the boom. The measures of total coliform outside the boom were statistically significantly higher in each case when each of the three paired sites is compared separately -- t=2.66 for Site 2 vs. Site 1; t=4.68 for Site 3 vs. Site 4; and t=2.67 for Site 5 vs. Site 6. As would be expected, since the individual comparisons were significant, when the measures of TABULAR DATA OMITTED TABULAR DATA OMITTED total coliform for all the outside sites (Sites 1, 4 and 6) were combined and compared to the combined inside sites (Sites 2, 3 and 5), the total coliform was also statistically significantly higher outside the boom (t=5.40).
The measures of fecal coliform outside the boom were statistically significantly higher in all three matched pairs (t=2.01 for Site 2 vs. Site 1; t=4.25 for Site 3 vs. Site 4; and t=2.17 for Site 5 vs. Site 6). Again, as expected, when the measures of fecal coliform for all the outside sites (Sites 1, 4 and 6) were combined and compared to the combined inside sites (Sites 2, 3 and 5), the fecal coliform was statistically significantly higher outside the boom (t=4.67). Chloride samples
The mean of sample logs of chloride measurements are reported and compared in Table 3 for the period from June 22 to August 19, 1992, which was a portion of the period after boom installation. The mean of sample logs for chlorides inside the boom was 4.04, which is statistically significantly higher than the mean of 3.96 for the sample logs measured outside the boom (t=4.05).
Discussion of Results
Coliform samples prior to installation of the boom
Analyses of samples collected in the beach waters prior to the installation of the boom indicate that, at the time of sampling, there was no statistically significant difference in total or fecal coliform levels between the area of water later to be enclosed within the boom and the area of water which would be outside the boom. Overall, the spring season samples were higher in both total and fecal coliform than the samples collected later in the summer. This trend is one that is usually observed and is believed to be due to heavy spring rains and storm water run-off that carry especially high loads of coliform into the harbor.
Coliform samples following installation of the boom
After boom installation, all logarithmic means for each of the three inside-the-boom sampling sites were lower in both total anti fecal coliform counts than their matched site outside the boom. Statistical significance was observed in all separate comparisons. Samples inside the boom were closer to code standard more often than samples outside the boom. For example, although the overall outside mean for the season was 19% below the maximum coliform count allowed by the code standard for a given 30-day period, it is evident that this water would have exceeded the standards at various times in the season. Moreover, the overall inside mean for the season was 69% below the 30-day code standard.
The logarithmic mean of total chlorides was slightly higher inside the boom than in waters outside the boom, for the period in which samples were collected. This is in direct opposition to the Nassau study, which found water within the boom to be 40% to 50% lower in salinity than surrounding waters. However, the Nassau study involved a smaller number of chloride samples and sampling in the center of the beach area, so direct comparisons of these results may not be appropriate.
One possible explanation for the higher chloride levels inside the boom is obvious, and, therefore, deserves brief discussion here. Previously collected data from Mamaroneck Harbor have shown that fresh water entering the harbor stays close to the surface. Based on historical data, we assume that this fresh water typically carried high levels of bacteria into harbor waters. From our study during the summer of 1992, we know that coliform levels within the boom were generally lower than those measured outside the barrier.
Table 3. Comparison of matched site means of paired sample measures of total chlorides after installation of the fabric boom Sites 2,3&5 vs. Sites 1,4&6 Sites 2,3&5 Sites 1,4&6 Variable Inside Outside Total Chlorides Number of paired samples 77 Mean of the sample logs 4.04 3.96(*) t=4.05 Antilogarithm of the mean of sample logs 10965 9120 * p|is less than~.05
All this suggests that the boom may be somewhat more effective at diverting surface water than subsurface water. According to this theory, the boom was able to divert fresh water from Mamaroneck River and Beaver Swamp Brook from the beach, in turn diverting the coliform that water carried, but as a result also increased the relative salinity of the water inside the boom (by reducing the opportunity for dilution by fresh water). However, detailed evaluation of this theory is beyond the scope of this study and additional work would be needed to confirm this.
Conclusion and Recommendations
The results of this study suggest that the boom was a factor in reducing total and fecal coliform bacteria during the 1992 bathing season. The logarithmic mean for total coliform for all of the samples from within the boom was 62% less than the outside samples. Similarly, the fecal coliform level was 52% less for the water inside the boom versus the outside waters. This is consistent with Nassau's findings of the boom used at Sea Cliff Beach in 1990, in which water within the boom proved to be lower in coliform and fecal coliform than the surrounding waters. The Village of Mamaroneck appears to have succeeded in its goal of providing its residents with cleaner bathing water than what would have otherwise been available during the summer of 1992. However, some cautions are warranted. The results and conclusion of this study should not be interpreted as an unequivocal endorsement for the use of such booms in all beaches desiring to reduce coliform bacteria in their bathing waters, or even in the Mamaroneck Harbor Beach in future seasons. The intricate and interdependent conditions that comprise the aquatic environment of beaches vary from beach to beach, and even from season to season within the same beach. Considerably more study is needed before any wholesale conclusion about the worth of such booms can be reached.
The limited nature of this study invites speculation as to the reasons behind our results. Aside from sampling only one bathing season, this study did not consider other important variables such as tide changes, weather conditions, and beach attendance. Nor was the WCDH study coordinated with any systematic sampling of the waters from the Beaver Swamp Brook and the Mamaroneck River, both suspected to be significant contributors of coliform to the harbor waters. Other communities have studied the correlation between water quality of bathing beaches and the inflow from tributaries and suggested that a watershed protection action plan needs to be considered to assure water quality in community bathing areas (7).
A future and more comprehensive study should control for one or all of these variables, for a fuller understanding of the boom and its effect in water quality. Such a study could assist communities in considering all alternatives to protect the quality of their bathing beach waters.
1. Stevenson, A.H. (1953), Studies of bathing water quality and health, J. Amer. Pub. Health Assoc. 43:529-533.
2. Chasis, S. and A. McLain (1993), Testing the Waters III: Closings, Costs and Cleanup at U.S. Beaches, Natural Resources Defense Council, New York, NY.
3. ... (1993), Pollution woes can leave beachgoers high and dry, Gannett Suburban Newspapers, July 1 ed., White Plains, NY.
4. ... (1991), Mamaroneck Harbor Project: Data Evaluation and Model Analysis, EPA Contract No. 68-C8-0105, U.S. Environmental Protection Agency, Office of Marine and Estuarine Protection, Washington, DC.
5. ... (1992), Chapter I State Sanitary Code Subpart 6-2 Bathing Beaches (Statutory Authority: Public Health Law, Sect. 225), (including amendments effective Oct. 7, 1992), New York State Dept. of Health, Bureau of Community Sanitation and Food Protection, Albany, NY.
6. ... (1992), Effect of a curtain boom on coliform bacteria levels at Sea Cliff Village Beach, In: Nassau County Dept. of Health Annual Report, city, NY. 7. Ingram, T.I. (1993), A preliminary investigation into the bacteriological water quality problems of Stonelick Lake State Park, Ohio, J. Env. Health 55(6): 19-23.
Dena Fisher, Ph.D., Asst. Commissioner, Planning & Evaluation, Westchester County Dept. of Health, 19 Bradhurst Ave., Hawthorne, NY 10532.
|Printer friendly Cite/link Email Feedback|
|Publication:||Journal of Environmental Health|
|Article Type:||Cover Story|
|Date:||Mar 1, 1994|
|Previous Article:||In the absence of understanding, perceptions are realities - Part II.|
|Next Article:||Groundwater site characterization: a systems perspective.|